▣ DATA BRIEF · SCHOOL SAFETY
☀ Key Stats ☀
◆ At-school theft fell from 14.1 victimizations per 1,000 students in 2014 to 5.2 per 1,000 in 2024, a decline of about 63%.
◆ Nonfatal violent victimization at school was 18.9 per 1,000 in 2014 and 21.8 per 1,000 in 2024, a difference that was not statistically significant.
◆ Overall at-school victimization declined from 33.0 to 27.0 per 1,000 over the decade, about 18%, but that change was also not statistically significant.
◆ In 2024, students experienced 27.0 victimizations per 1,000 at school, compared with 17.7 per 1,000 away from school.
◆ Violent victimization at school rose from 14.2 per 1,000 in 2023 to 21.8 per 1,000 in 2024, a statistically significant increase at the 90% confidence level.
◆ Students from households earning under $25,000 had an at-school victimization rate of 60.3 per 1,000, versus 26.7 per 1,000 among those in households earning $25,000 to $49,999.
◆ At-school theft was higher among students ages 12 to 14 at 7.5 per 1,000, compared with 3.0 per 1,000 among students ages 15 to 18.
Continue reading ↓
Theft victimization at school among U.S. students ages 12 to 18 fell by about 63% between 2014 and 2024, while violent victimization showed no statistically significant decline over the same period, according to new Bureau of Justice Statistics data.
The at-school theft rate dropped from 14.1 victimizations per 1,000 students in 2014 to 5.2 per 1,000 in 2024.
Nonfatal violent victimization moved in the opposite numerical direction, from 18.9 to 21.8 per 1,000, but the Bureau of Justice Statistics found that the decade-long change was not statistically significant.
As a result, the total at-school victimization rate fell by a much smaller 18%, from 33.0 to 27.0 per 1,000, and that overall decline was also not statistically significant.
◎ HEADLINE CHANGE
−63%
Decline in the at-school theft victimization rate between 2014 and 2024, from 14.1 to 5.2 victimizations per 1,000 students.
School theft fell sharply over the decade
The decline in theft was one of the clearest long-term changes in the federal data.
At-school theft fell from 14.1 per 1,000 students in 2014 to 5.2 per 1,000 in 2024.
That works out to a decline of approximately 63%.
Theft away from school also declined substantially, from 11.2 to 5.1 victimizations per 1,000 students, a reduction of roughly 54%.
BJS found both decade-long declines in theft to be statistically significant.
The measure includes attempted and completed purse snatching, completed pocket picking and most other attempted or completed theft, but excludes motor vehicle theft and robbery.
↳ VIOLENCE · NO SIGNIFICANT DECADE-LONG DECLINE
Violent victimization remained statistically unchanged from 2014
Nonfatal violent victimization followed a different pattern.
The at-school rate was 18.9 per 1,000 students in 2014 and 21.8 per 1,000 in 2024.
Although the 2024 rate was numerically about 15% higher, BJS found no statistically significant difference between the two years.
Nonfatal violent victimization includes rape or sexual assault, robbery, aggravated assault and simple assault.
When simple assault was excluded, the at-school rate was 5.8 per 1,000 in 2024, compared with 3.6 per 1,000 in 2014. That decade-long difference was also not statistically significant.
↔ 2014 TO 2024 AT SCHOOL
Theft · 14.1 → 5.2 per 1,000
Violent victimization · 18.9 → 21.8 per 1,000
Total victimization · 33.0 → 27.0 per 1,000
Violent victimization increased between 2023 and 2024
The longer-term stability masks a more recent increase.
At-school violent victimization rose from 14.2 per 1,000 students in 2023 to 21.8 per 1,000 in 2024, an increase of about 54%.
BJS reports that difference as statistically significant at the 90% confidence level.
However, the more serious category that excludes simple assault did not change significantly between 2023 and 2024.
! IMPORTANT DISTINCTION
A one-year increase can coexist with no statistically significant decade-long trend. The 2024 violent-victimization estimate also has a larger margin of error than many earlier annual estimates.
↔ LOCATION · AT-SCHOOL RATE WAS HIGHER
Students experienced more victimization at school than away from school
Across all theft and nonfatal violent victimization combined, the rate at school was 27.0 per 1,000 students in 2024.
That compared with 17.7 per 1,000 away from school, making the at-school rate about 53% higher.
BJS found the difference between the two locations statistically significant.
For the report, “at school” includes incidents inside school buildings, elsewhere on school property, on school buses and while students were traveling to or from school.
◎ LOCATION COMPARISON
27.0 vs. 17.7 per 1,000
Overall victimization was about 53% higher at school than away from school in 2024.
Lower-income students had a much higher at-school victimization rate
Victimization also differed by household income.
Students living in households earning less than $25,000 per year had an at-school total victimization rate of 60.3 per 1,000.
That was more than twice the 26.7 per 1,000 rate among students from households earning between $25,000 and $49,999.
The under-$25,000 estimate was also more than three times the 18.9 per 1,000 rate among students from households earning $100,000 or more, although BJS’s formal statistical comparisons in the report use the $25,000-to-$49,999 group as the reference category.
↳ AGE · YOUNGER STUDENTS SAW MORE THEFT
School theft was 2.5 times as high among students ages 12 to 14
Students ages 12 to 14 experienced an at-school theft rate of 7.5 per 1,000 in 2024.
Among students ages 15 to 18, the rate was 3.0 per 1,000.
That makes the younger group’s estimated theft rate 2.5 times as high, and BJS found the difference statistically significant at the 90% confidence level.
Total at-school victimization, however, did not differ significantly between the two age groups.
Male and female students had similar victimization rates at school
BJS found no statistically significant differences between male and female students in any of the at-school crime categories measured in 2024.
Away from school, however, total victimization was higher for female students at 23.5 per 1,000, compared with 12.2 per 1,000 for male students.
The difference was statistically significant at the 90% confidence level.
✦ Why it matters ✦
The decade-long fall in theft shows that one important source of student victimization has declined substantially.
But the broader school-safety picture is more mixed because violent victimization has not shown the same sustained reduction.
That distinction also explains why the total school victimization rate fell only about 18% from 2014 to 2024 even as theft dropped 63%.
The income and age differences further suggest that national averages can conceal substantial variation in which students experience victimization.
The latest figures therefore point to significant progress on school theft without equivalent evidence of a decade-long improvement in nonfatal violent victimization.
ⓘ How to read the findings
The figures come from the Bureau of Justice Statistics’ National Crime Victimization Survey, a nationally representative survey in which people report criminal victimization they experienced. The statistics are not limited to incidents reported to police.
The report covers students ages 12 to 18 whose educational attainment had not exceeded grade 12 at the time of the survey.
“At school” includes incidents in school buildings, on school property, on school buses and while going to or from school. It therefore covers a broader setting than incidents occurring only inside classrooms or school buildings.
Total victimization combines theft and nonfatal violent victimization. Nonfatal violence includes rape or sexual assault, robbery, aggravated assault and simple assault.
The 63% theft decline is calculated from the reported rates of 14.1 per 1,000 in 2014 and 5.2 per 1,000 in 2024. BJS reports that difference as statistically significant.
The numerical increase in violent victimization from 18.9 to 21.8 per 1,000 between 2014 and 2024 was not statistically significant. The data therefore do not establish that school violence increased over the decade.
The 2024 NCVS transitioned to a redesigned survey instrument using a split-sample approach. This report uses respondents who received the legacy questionnaire so that comparisons with earlier years remain possible, but dividing the sample reduced the precision of 2024 estimates.
BJS reports significance at both the 90% and 95% confidence levels. Some subgroup estimates in the underlying tables are flagged for caution because they rely on small numbers of sample cases or have high coefficients of variation; those estimates should not be treated as equally precise as the national totals.
Due to methodological changes in the 2016 NCVS data, youth victimization estimates for that year are not comparable with the other years and are omitted from the trend figures.
